High-intensity sprints of short duration, interspersed with brief recoveries, are common during most team sports. Therefore, the ability to recover and to reproduce performance in subsequent sprints is an important fitness requirement of team-sport athletes and has been termed repeated-sprint ability (RSA). Using laboratory and field-based protocols, performance analyses have shown that fatigue during repeated sprints is manifested by decreased running speed or reduced power output. Proposed factors responsible for these performance decrements include limitations in energy supply, metabolic by-product accumulation (e.g., P_i, H⁺), impairment of Ca²⁺ kinetics or reduced excitation of the sarcolemma (increase in extra-cellular K⁺). However, increasing evidence also indicates that failure to fully activate the contracting muscle might potentially limit repeated-sprint performance. Moreover, the details of the task (e.g., changes in the nature of the work/recovery bouts) will determine the contribution of the underlying mechanisms (task dependency) and additional homeostatic perturbations (e.g., hypoglycaemia, muscle damages, hyperthermia) are likely to compromise fatigue resistance. It could be argued that the best way to improve repeated-sprint performance is to improve the underlying factors responsible for fatigue. High-intensity interval training (80-90% VO_2max) interspersed with rest periods that are shorter than the work periods appear necessary to increase the ability to buffer H⁺ (βm). We have also shown that similar high-intensity interval training can significantly improve PCr resynthesis and RSA. In contrast, intermittent sprint training has not been reported to increase βm or the rate of PCr resynthesis. Thus, rather than game-specific training, training specific for the underlying physiological systems may be required to improve these important physiological qualities for team-sport athletes. While a single sprint relies predominately on anaerobic energy sources, recent research has shown that it is possible for athletes to reach their VO_2max during repeated sprints. This suggests that training that increases VO_2max may allow for a greater aerobic contribution during latter sprints, potentially improving repeated-sprint performance. In summary, this chapter attempts to summarise and draw conclusions from the studies that have described mechanisms underlying functional impairments during repeated-sprint tasks and to relate these decrements to the potential impact of various training regimens on repeated-sprint performance.